Tuesday, February 16, 2010

SMBG Research, Or The Lack Of It


Nearly three years ago I challenged the authors of a particularly poor paper titled "Impact of self monitoring of blood glucose (SMBG) in the management of patients with non-insulin treated diabetes: open parallel group randomised trial" in the British Medical Journal to conduct a study based on Test, Review, Adjust.

When I first read Jennifer's Test, Test, Test I used to wonder why a study of that method had never been performed.

Of course now, with a little more experience, the reasons are fairly obvious to me. To start with, no researchers are even vaguely aware of the technique. That is apart from the fact that the results are unlikely to lead to increases in medication sales. Over five years ago, back when I naively thought someone would listen, I rang the Australian offices of Roche and Lilly to suggest it. They were polite and totally uninterested. I thought that at least the major test-strip manufacturers may have a vested interest. Apparently not.

My response to that paper was another small attempt to interest somebody in the appropriate field in the concept. But I have to be realistic; not many researchers read the "Rapid Responses" and Farmer et al certainly were not going make any attempt to prove themselves wrong.

The idea surfaced again today when Stuart, a Type 1 diabetic on the dLife forum, posted this very interesting question:

"If you wanted someone to explore something, to do a STUDY on a subject(s) about our diabetes, what do you want them to research? What areas do YOU want to know far more about that don't seem to be being done??? "

He had some very interesting replies. You can read them here. I would like to expand slightly on my answer there.

There are so many areas of diabetes crying out for research. There are some that have never been studied at all, including those dealing with diet modified by structured testing or similar methods which can lead to minimal medication or insulin needs. At the moment research tends to be focused on finding new medications or new ways to use old medications. In the real world "who pays the piper calls the tune".

No researchers are asking us, the diabetics, "what should we be researching?"

My own area of interest is just one of many possibilities. Despite understanding the reasons I still find it hard to believe that after more than three decades of home self-testing of blood glucose by diabetics no medical researcher, anywhere, has researched the use of structured self-testing for dietary modification to reduce blood glucose excursions.

Thousands of type 2 diabetics like myself have been "researching" the method personally and reporting their individual successes on many different forums since before I was diagnosed eight years ago, but we don't count in professional medical and research terms. We are diabetics, not scientists and our reports are anecdotes, not data.

I will offer the basic idea. Who knows, maybe there is a bright scientist out there looking for a PhD subject who has the ability to find a grant or research funds.

I propose a study comparing two groups of type 2 diabetics, all within their first 12 months of diagnosis. The only exclusion criteria would be that none should be using insulin or an insulin-stimulating medication such as a sulfonylurea at the commencement of the study.

Group 1, control, would be treated as individuals by their physicians and other specialists in exactly the same way as the present guidelines for their country. For example that would be the ADA or AACE and American Dietetic Association guidelines in the USA, Diabetes Australia here, or the NICE/NHS guidelines in the UK.

Group 2 would would also be treated as individuals by their physicians and other specialists in exactly the same way as the present guidelines for their country with the exception of dietary and testing guidelines. Instead, they would be given basic dietary guidelines to understand the differences between carbohydrates, fats and protein and their effect on blood glucose levels, and would also receive training and support in using feedback from peak post-prandial blood glucose testing to modify diet and lifestyle. The method taught would be based on the technique described in Test, Review, Adjust. If that needs clarification I am available as a consultant :)

The period of study would be three months initially, with weekly support and review to record indicators for both groups for the first four weeks, then monthly for the next two months with a preliminary report prepared after three months. Periodic follow-up review and reports would be performed at 12 months, five years and ten years. Indicators recorded would include A1c, fasting and peak post-prandial blood glucose levels, lipids, weight, blood pressure and any others the researchers felt valuable.

The five year and ten year reports would follow up all the earlier results and also include morbidity and mortality and any differences in progression to, or of, diabetes complications.

An inexpensive pilot study would not need very large populations and could be restricted to the first three months. The results of that could support further study over the longer period with a larger population.

I can also see other possible studies. For example, the possibility of combining Gannon and Nuttall's LoBAG20 or LoBAG30 diet with the above study as the starting diet for Group 2 is one that intrigues me. But it may be unwise to put too many variables in the mix. One thing at a time.

My area of interest may be quite different to yours. If you were the person asked by the researchers "what do you want us to study?" what would your answer be?

Cheers, Alan

Everything in Moderation - Except Laughter

11 comments:

Jenny said...

I'd like to see some research about the 10 year complication outcomes for people with Type 2 who control by cutting carbs. With break outs of the outcome in the range below 6.5% A1c.

There isn't a single study that looks at the effect of sustaining control under the 6% A1c with diet. Not one.

That's because the ONLY diet that does it involves cutting carbs and does NOT use expensive drugs.

Follow the funding on most studies and you'll find a drug company looking for more data it can use to sell drugs that cost US$180/month.

Effective dietary control takes money out of drug company pockets. So you won't see them ever fund those studies.

Alan said...

Thanks Jenny

You know I agree, of course. I suppose the only possibility is a study funded by a private grant or a government. Well, we can live in hope.

I've been drafting a comment to your CF7L2 post (my brain just can't remember all those genome numbers) trying to not look too dumb.

Basically I reckon type 2 is a family of conditions including many combinations of of beta cell depletion, insulin resistance, signalling flaws and several other possible components which can kick in at different times along the progression path. But it's past midnight here so that will have to wait for the morning when I may find a way to express it sensibly:)

Margie's Musings said...

Are you suggesting using only diet to control glucose, Alan? Bob uses Metformin twice a day, plus watching his carbs.

It's tragic that capitalism encourages the kind of greed displayed by the drug companies.

Alan said...

Specifically for the research study that I proposed, I suggest excluding patients who are using insulin or insulin-stimulating medications. Not because "test, review, adjust" may not work for them, but because the medication may invalidate research comparisons.

For individuals, I am not anti-medication. I added metformin myself after using just diet and exercise for my first three years.

If medication is needed I believe "just enough" medication or insulin, to complement a way of eating that does not cause blood glucose spikes, is a better and safer path to follow than using over-medication to counter inappropriate carb consumption.

Anonymous said...

I would like to see a study on whether gluten intolerance plays into the development of insulin resistance and diabetes (both types) and/or the impact on complications if gluten intolerance remains undiagnosed or ignored (as in someone who cuts down on their carbs but is unable to resist wheat/gluten products.

I say this because I cut bread, etc. completely out of my diet and got down to a 5.5 A1C after discontinuing my medication. My 2 sisters, on the other hand, refuse to give up the bread (including white bread - ugh) and pasta. Both still have above 6 A1C's and are on meds (one is on 2 meds - the heavy bread eater).

From what I've read about gluten intolerance, it can cause major damage to the digestive system, blocking intake of important nutrients. Perhaps the first foray into cereals as a baby could start the ball rolling on developing damage, including triggering the genetic tendencies toward diabetes - both types. Perhaps the explanation for Type 1 in very young children.

Just wondering is all...

Alan said...

Mayne it was gluten intolerance for you. Maybe not.

Unless you suffer from coeliac disease or something similar I suspect your improvement had more to do with the reduction in carbohydrates than a reduction in gluten.

Gys de Jongh said...

Hi Alan,
nice proposal. In the mean time the first experiment along those lines have been done. Thought you might want to know :)

the first article, I know off, that uses the feedback from the blood glucose meter and a lot of knowledge to design your own woe

http://www.ncbi.nlm.nih.gov/pubmed/19934460?dopt=Citation
Diabetes Educ. 2009 Nov-Dec;35(6):1023-30.
Using meal-based self-monitoring blood glucose (SMBG) data to guide dietary recommendations in patients with diabetes.

The purpose of this article is to describe how self-monitoring of blood glucose (SMBG) data is a useful tool for identifying and managing postprandial hyperglycemia (PPHG). PPHG and postprandial glucose excursions occur frequently in patients with diabetes even when hemoglobin A1C is controlled below 7.0%, and convey increased risk of cardiovascular morbidity and mortality. Consequently, effective management of diabetes must include control of postprandial glucose levels. Postprandial plasma glucose (PPG) depends on the composition of meals, specifically the amount of carbohydrates. Reduced-carbohydrate diets offer short-term improvements in glycemic control and other metabolic parameters, but await the support of long-term efficacy and safety studies. Glucose profiling and paired-meal SMBG are useful tools for detecting PPHG and glucose excursions. They provide immediate feedback to patients on the effect of foods and meals, thereby allowing appropriate food and medication adjustments to improve postprandial glycemic control.

PMID: 19934460

Alan said...

Gys, thank you so much.

You just gave me a lovely birthday present. I have been waiting too lontg to see a study like this happening.

I was hoping someone would know of a study. Of course, I should have realised that if anyone would, you would.

For this one I may even open the wallet and pay for the full text.

Great to hear from you Gys; I'll be in touch.

Alan said...

Someday I'll discover how to edit comments to fix my typos :-)

Anonymous said...

Hi everyone! I don't know where to start but hope this site will be useful for me.
Hope to receive some help from you if I will have some quesitons.
Thanks in advance and good luck! :)

trinkwasser said...

A significant number of diabetics of both (all) types and also nondiabetics have problems with wheat over and above gluten intolerance. It's pretty toxic even compared to other grains with its lectins, phytates and wheat germ agglutinin which many of us simply haven't had time to adapt to even if we've evolved to deal with the gluten.

Current world dietary policies may be an attempt to breed wheat intolerance out of the population. This would appear to take precedence over researching ways to reduce the harm from diabetes.

I've been thinking, if anyone would be able to garner finance for such a study it would be the Metabolism Society

http://www.metabolismsociety.org/Default.aspx

I'd been thinking of writing to suggest a study of Test Test Test. Now I'm going to do just that, maybe y'all would like to get on board. I got bogged down in reading some of the other appalling non-studies as examples of how not to do it.